TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Ask HN: Would you pay for a verified-only social network?

11 点作者 bayeslaw大约 2 年前
We are building an MVP for a new kind of social network (I know, don&#x27;t yawn just yet).<p>Here&#x27;s our view:<p>GPT4 will transform the internet in many ways. One of which is the trust we have in our social interactions online.<p>How can you know, that you&#x27;re not talking to (or arguing with at 2am) with a bot, trying to sell you something or recruit you for a cause? In a few months you simply won&#x27;t be able to tell the difference.<p>This, on top of the well known pitfalls of existing social networks (gaming our attention with divisive content to sell ads to us) made us think: what if we had a social network with the following properties?<p>- All users are ID verified. - All users pay a monthly (1$ or so) fee.<p>These two conditions would already result in:<p>- No need for ads or gaming your attention or selling your data. - No bot generated content. - Healthier more humane discussions.<p>We think, there&#x27;s a huge qualitative difference between having ALL your users verified vs only some of them.<p>Thoughts?

28 条评论

codingdave大约 2 年前
I have zero interest in this, and would not pay.<p>It is not a bad idea, but the problem with this is that social media is not a requirement in my life. Some people enjoy it, so you will probably find a market, but many people simply have lost interest in the entire concept of social media, and I am in that group.<p>You are on track to be asking about the interest level is in this, and I&#x27;d be sure to ask this question to a wide variety of demographics to understand exactly how much market potential really exists.
rainytuesday大约 2 年前
Nextdoor sort of does that. They charge you $1 to verify your mailing address --then you get to post in your neighborhood group if you use your real name. Doesn&#x27;t seem to have turned out all that great.
Guest9081239812大约 2 年前
A few points...<p>1. You&#x27;re saying we&#x27;ll not be able to distinguish between human and bot content in a few months. If that&#x27;s true, then what stops someone from signing up for $1 with their ID and having a bot write all their content? I could easily have a bot write hourly messages, self-promotions and content about my business. The only thing you can guarantee is someone is paying a fee. That&#x27;s it.<p>2. You said you want to use images for identification. What happens when you cannot distinguish a real person from an AI person, which seems to already exist (<a href="https:&#x2F;&#x2F;twitter.com&#x2F;mileszim&#x2F;status&#x2F;1613965684937224192" rel="nofollow">https:&#x2F;&#x2F;twitter.com&#x2F;mileszim&#x2F;status&#x2F;1613965684937224192</a>)? You said in another comment the person needs to be in their photo with the ID. I Photoshopped one in for you (<a href="https:&#x2F;&#x2F;i.imgur.com&#x2F;42ip7Ih.jpg" rel="nofollow">https:&#x2F;&#x2F;i.imgur.com&#x2F;42ip7Ih.jpg</a>) and now we have one fake AI generated human with a Washington driver&#x27;s license. Yes, the photo on the license doesn&#x27;t match and the hand is an unrealistic position, but I could fix that with 15 minutes of editing and then reuse the hands to overlay on other fake accounts.<p>3. It seems likely that bot content becomes higher quality than human content. Let&#x27;s be honest here, human written content is full of ulterior motives, deceit, manipulation, marketing, scams, etc. Wouldn&#x27;t I be better off having a debate with a bot that is more reasonable, more eloquent, and more knowledgeable? Is having a 2AM argument with a human on a social network a valuable experience I should be paying for? It&#x27;s one side yelling at the other and it never goes anywhere productive. You could be going the wrong direction here. What about a social network that&#x27;s 80% content written by friendly, happy, AI bots that you provide? They build a positive atmosphere and strong foundation for the 20% of humans to stand on. They push the conversation in a better direction, they bring in more facts, they compliment people and make them feel better about themselves and their appearance. Why can&#x27;t we use bots to our advantage? Surround people with happy bots, and maybe they, themselves will be more likely to write thoughtful, respectful messages.
CM30大约 2 年前
No, I wouldn&#x27;t pay for it. And there are multiple reasons for that:<p>1. Social media services don&#x27;t bring enough value to be worth paying for, since it&#x27;s unlikely most others will do the same. So if I pay for your social network, I&#x27;m likely losing contact with the folks I care about (at least there) in the vague hope I&#x27;ll find new people to talk to. That&#x27;s rare enough with free services, let alone paid ones.<p>2. Similarly, by being paid, it&#x27;s less likely the communities and people I&#x27;d want to talk to would be on your site&#x2F;service. Even with machine generated content, there&#x27;s more interesting stuff on Twitter&#x2F;Reddit&#x2F;YouTube&#x2F;whatever than this hypothetical paid one, so I&#x27;d rather take my chances on the larger sites.<p>3. The more people want others to hear what they say and pay for the privilege, the less others are likely to want to hear from them. See Twitter Blue and the kinds of people who tend to pay for it. Your service could end up being Twitter Blue the Social Network, with all the vapid &#x27;thought leaders&#x27; and &#x27;would be influencers&#x27; that brings.<p>4. I really don&#x27;t want to use a service that requires you participate under your &#x27;real life&#x27; identity, since that usually leads to lesss interesting discussions and communities (with very little effect on toxicity, as Facebook shows).<p>5. I certainly don&#x27;t want a social media company having proof of my ID, both for privacy reasons (this data could be sold to other companies) and safety ones (like if a Musk style figure takes over, or an authoritarian regime comes into power).
hammyhavoc大约 2 年前
Anybody remember Path? <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Path_(social_network)" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Path_(social_network)</a><p>And the answer is no. You aren&#x27;t putting the decentralized genie back in the bottle for anybody that still cares about social media.<p>With 100,000 tweets and a few tens of thousands of followers, I was terminally online for many years. I&#x27;m now much happier that I don&#x27;t use it. Echo chambers are crap, and for anyone I want to talk to, I do.
rg111大约 2 年前
No need to get disheartened for the all the comments here. I am sure there is a market for this.<p>I personally wouldn’t use it, because I have two kinds of persistent online circles: one for close friends and family. We talk via encrypted messaging, video calls, etc. Another is online study groups, language&#x2F;framework forums, interest groups. Don&#x27;t see the need of verification here. We meet face to face or over video call&#x2F;chat. In such places, every active member knows other active members.<p>I just am disgusted by Facebook, LinkedIn, etc. I am there only because local bakeries, butchers, and laundries are there. Some acquaintances with whom I don&#x27;t like to share my number are there, and I can reach them via messaging.<p>If you add a payment barrier, most people won&#x27;t be there. And so local bakeries, butchers, and laundries won&#x27;t be there. Why would I use it?<p>If you already have achieved something like FB (who am I kidding?), and I can&#x27;t ignore your service, sure, I will pay for it.<p>And, if a GPT-4 bot is talking sense, I will chat with it in the public. Others can learn from it, or I can learn something from it- factual things.<p>And if I don&#x27;t like a bot&#x27;s line of talking- I will just block (or kick out) and move on.
hayst4ck大约 2 年前
I think a well executed walled garden social network could succeed. Reddit is chasing the lowest common denominator, and something that chases higher brow content could be good. Something that tries to be more hard-lined about bad faith speech&#x2F;users could probably do well.<p>AI dang? AI automoderating away calling someone names. AI weighing in on factual correctness? AI tagging? All things that might add value to an online forum. A lot of forums are post based, but I imagine that altering content ranking based on reputation might be interesting.<p>I&#x27;m not sure you have the right approach and until you have a network of users you have no value, so I&#x27;m not sure why anyone would pay.<p>I don&#x27;t think you&#x27;ve thought about your very real bootstapping problem enough. I think your bootstrapping problem is big enough that it dominates all other considerations. What&#x27;s going to make people come to your site? After you have an answer to that, you are probably going to pivot away from what you think you&#x27;re building right now anyway.
fogzen大约 2 年前
Yes, but no. Yes I will pay to verify my identity. But no, I want to control my identity: My own domain. My email is “verified” (I own the DNS) but I can move to another provider while keeping my email. Public key identities work the same way (like what Bluesky does).<p>Free + ads is a <i>great</i> business model. It works. There’s a reason all major consumer platforms work that way. The world, in general, is not ideologically hungup about ads like HN is. Also notice that most of HN‘s rhetoric doesn’t match their actions, most are using popular social media anyway.
thinkingemote大约 2 年前
Technically: Human verification online will be very difficult making this solution hard.<p>video, audio, chat will all be considered &quot;sus&quot;<p>I think that this change you are responding to will hopefully lead to more IRL meetups like there was pre social media.<p>Given this I think Facebook remains and will benefit. It has already a network of people who know each other and it has good IRL meeting support. So you might be competing with them.<p>The one thing I can think of is using face to face in world meetings as a basis for the verification and network creation. Look up pgp signing parties.
zepearl大约 2 年前
&gt; <i>All users are ID verified. - All users pay a monthly (1$ or so) fee</i><p>You probably mean the following?<p>&gt; <i>All users who are allowed to write&#x2F;post are ID verified - All users who are allowed to write&#x2F;post pay a monthly (1$ or so) fee&quot;</i><p>In general I tend to want to warn you: ID verification is probably not a cheap thing to do (time-consuming, very difficult to provide an absolute &quot;guarantee&quot;, refer to checks done to open a bank account).
cookiengineer大约 2 年前
Disclaimer: European citizen.<p>If you are based in the US then I&#x27;d say nope. This is pretty much identical as to the value proposition of all those paid dating platforms.<p>And we all know how that turned out, only scammer are on those platforms and the term &quot;love scam&quot; is a real thing these days because of it.<p>There&#x27;s lots of documentaries on the problem, how they hire people to keep their paying users busy etc.
jWhick大约 2 年前
I would, I honestly i had an idea the other day where it would be good if social media would limit the posts per day at 1 and have fixed timeline (show everything in chronological order, as opposed to currated algorithm feed). It would block people spamming for attention&#x2F;exposure.
danwee大约 2 年前
Shouldn&#x27;t it be the other way around? A &quot;verified-only&quot; social network should pay me if they want to get my real id&#x2F;passport. There&#x27; no chance I&#x27;ll give so much information to a social network company.
h2odragon大约 2 年前
fark.com tried something like that<p>Remember when Musk said he was buying Twitter to solve the problem of Authenticating the World? This may have been what he was talking about.<p>I think the widespread reluctance to talk much about the concept stems from the power given to the Authentication Authority, whoever they wind up being. They become the arbiter of who or what can participate in society almost instantly. We don&#x27;t have anyone we trust that much.
rd大约 2 年前
I would. I would also just continue using Twitter if I could pay more than 8$&#x2F;mo and get NO ads. No ads is really all I care about.
billconan大约 2 年前
I will not use this. This sounds like another facebook.<p>(At least from where I come from), we can only speak the truth when our identities are fake.
2rsf大约 2 年前
No. I don&#x27;t see the need, at least not yet.<p>What&#x27;s the business model if there is no advertisement and only 1$ subscription?
评论 #35826262 未加载
Am4TIfIsER0ppos大约 2 年前
Assuming you manage any of that I have a question: will the posts be anonymous or pseudonymous to all the other users?
评论 #35817838 未加载
throwawayadvsec大约 2 年前
&quot;- All users are ID verified. - All users pay a monthly (1$ or so) fee.&quot;<p>That will absolutely NOT stop bots.
评论 #35821753 未加载
sergiotapia大约 2 年前
No. I value anonymity online. I only use my real name here because HN is more like a linkedin for me.
xtrh大约 2 年前
How about have monthly fee but no ID verify?
p1esk大约 2 年前
No.
jazzyjackson大约 2 年前
you don&#x27;t get to know my home address, sorry
评论 #35826283 未加载
juliangmp大约 2 年前
While I can see where youre coming from – misinformation is already a problem and modern AI will make it a lot worse, I dont think I would want to give up my anonymity for that. Even if my real name wasn&#x27;t publicly linked to my account.
endorphine大约 2 年前
No.
naru_s大约 2 年前
No
cloudking大约 2 年前
Do it, call it Human
forgotmysn大约 2 年前
absolutely not